Search Results (1 - 10 of 10 Results)

Sort By  
Sort Dir
 
Results per page  

Abounia Omran, BehzadApplication of Data Mining and Big Data Analytics in the Construction Industry
Doctor of Philosophy, The Ohio State University, 2016, Food, Agricultural and Biological Engineering
In recent years, the digital world has experienced an explosion in the magnitude of data being captured and recorded in various industry fields. Accordingly, big data management has emerged to analyze and extract value out of the collected data. The traditional construction industry is also experiencing an increase in data generation and storage. However, its potential and ability for adopting big data techniques have not been adequately studied. This research investigates the trends of utilizing big data techniques in the construction research community, which eventually will impact construction practice. For this purpose, the application of 26 popular big data analysis techniques in six different construction research areas (represented by 30 prestigious construction journals) was reviewed. Trends, applications, and their associations in each of the six research areas were analyzed. Then, a more in-depth analysis was performed for two of the research areas including construction project management and computation and analytics in construction to map the associations and trends between different construction research subjects and selected analytical techniques. In the next step, the results from trend and subject analysis were used to identify a promising technique, Artificial Neural Network (ANN), for studying two construction-related subjects, including prediction of concrete properties and prediction of soil erosion quantity in highway slopes. This research also compared the performance and applicability of ANN against eight predictive modeling techniques commonly used by other industries in predicting the compressive strength of environmentally friendly concrete. The results of this research provide a comprehensive analysis of the current status of applying big data analytics techniques in construction research, including trends, frequencies, and usage distribution in six different construction-related research areas, and demonstrate the applicability and performance level of selected data analytics techniques with an emphasis on ANN in construction-related studies. The main purpose of this dissertation was to help practitioners and researchers identify a suitable and applicable data analytics technique for their specific construction/research issue(s) or to provide insights into potential research directions.

Committee:

Qian Chen, Dr. (Advisor)

Subjects:

Civil Engineering; Comparative Literature; Computer Science

Keywords:

Construction Industry; Big Data; Data Analytics; Data mining; Artificial Neural Network; ANN; Compressive Strength; Environmentally Friendly Concrete; Soil Erosion; Highway Slope; Predictive Modeling; Comparative Analysis

Flessner, Brandon PSPECIES DISTRIBUTION MODELING OF AMERICAN BEECH (FAGUS GRANDIFOLIA EHRH.) DISTRIBUTION IN SOUTHWESTERN OHIO
Master of Arts, Miami University, 2014, Geography
The ability to predict American beech distribution (Fagus grandifolia Ehrh.) from environmental data was tested by using a geographic information system (GIS) in tandem with species distribution models (SDMs). The study was conducted in Butler and Preble counties in Ohio, USA. Topography, soils, and disturbance were approximated through 15 predictor variables with presence/absence and basal area serving as the response variables. Using a generalized linear model (GLM) and a boosted regression tree (BRT) model, curvature, elevation, and tasseled cap greenness were shown to be significant predictors of beech presence. Each of these variables was positively related to beech presence. A linear model using presence only data was not effective in predicting basal area due to a small sample size. This study demonstrates that SDMs can be used successfully to advance our understanding of the relationship between tree species presence and environmental factors. Large sample sizes are needed to successfully model continuous variables.

Committee:

Mary Henry, PhD (Advisor); David Gorchov, PhD (Committee Member); Jerry Green, PhD (Committee Member)

Subjects:

Botany; Ecology; Forestry; Geography

Keywords:

American beech; Fagus grandifolia; species distribution model; SDM; boosted regression tree; BRT; GIS; geographic information system; predictive modeling; Ohio; generalized linear model; GLM;

Uhrig, Lana KFeasibility of a long-term food-based prevention trial with black raspberries in a post-surgical oral cancer population: Adherence and modulation of biomarkers of DNA damage
Doctor of Philosophy, The Ohio State University, 2014, Public Health
Consumption of phytochemical-rich fruits and vegetables is associated with a lower risk of oral cancer and protection against oxidative stress-mediated damage from reactive oxygen species (ROS). It is proposed that antioxidant components in black raspberries (BRB) can “scavenge” ROS and diminish oxidative burden in the oral cavity. Therefore, we tested the hypothesis that a long-term, low-dose dietary administration of BRB to a population of disease-free oral cancer survivors is both (i) achievable and (ii) will result in the attenuation of oxidative DNA damage. Participants were assigned to consume 0, 4, or 8 grams of BRB daily for 6 months and provided self-report logbooks of adherence. Mass spectrometry of participant urine was used to identify biomarkers of adherence: dimethyl ellagic acid (DMEA) and urolithins. ELISA analysis of urine was used to measure 8-hydroxy-2’-deoxyguanosine (8-OHdG). Nanostring technology was employed to interrogate gene expression pathways associated with oxidative stress and DNA damage. Furthermore, a logistic regression model was developed to determine factors associated with willingness to participate and continue with study once enrolled. 112 participants were enrolled to a BRB or control regimen. Mean self-reported adherence to the study regimen for those returning logbooks was 88% at 10 & 20 weeks (CI: 83.87-92.65 & 85.81-91.89, respectively). MS/MS measurements of DMEA was 10-fold higher in BRB treated participants (p<0.0001 and 0.0015) at 10 and 20 weeks and urolithins were significantly higher in BRB participant urine at 10 weeks (p-0.0245). ELISA measurements of 8-OHdG in urine at 10 and 20 weeks was significantly decreased (p=0.0183 and p=0.0102) in BRB participants. Gene analysis of mucosal samples demonstrated significant down-regulation of NFKB2 (p=0.0282), NRF2 (0.0507), NQO1 (p=0.0011), GCLC (p=0.0468), and up-regulation of KEAP1 (p=0.0194) in BRB participants at 10 weeks. Predictive modeling revealed barriers associated with participation in the study which included: current smoker vs. never/past (OR: 7.85/4.86, p<0.0001/p=0.0003), physician engagement (OR: 5.71, p=0.0005), increasing age (p<0.0014), and adjuvant treatment (chemotherapy and/or radiation vs. none) (OR: 5.91 and 3.46, p<0.0001 and p=0.001). Once enrolled on study, those having the least travel distance (p<0.003), and government (OR: 3.96 & 7.04; p<0.02) or private insurance (OR: 5.1; p<0.01) vs. no insurance were more likely to complete. Additionally, past (OR: 3.85; p<0.02) and current smokers/tobacco users (OR: 3.82; p<0.03) continuing on after week 10 were more likely to complete than never smokers. This research demonstrates that a food-based prevention strategy, utilizing BRB in oral cancer survivors, is safe and feasible. Participants completing the study maintained an outstanding level of adherence with humoral biomarkers (DMEA, urolithins) serving as objective means of validation. Issues influencing the initial decision of patients to enroll and continue on once enrolled were identified. Long-term, low-dose administration of BRB effectively decreased urinary levels of 8-OHdG, an established biomarker of DNA damage, and significantly regulated genes/pathways associated with inflammation, carcinogen metabolism, and oxidative stress, all of which are associated with development of oral cancer. Results from this study provide evidence that administration of BRB may offer loco-regional and systemic benefits with very low potential for adverse effects.

Committee:

Christopher Weghorst, PhD (Advisor); Steven Clinton, MD, PhD (Committee Member); Dennis Pearl, PhD (Committee Member); Sun Qingua, MD, PhD (Committee Member); Randi Love, PhD (Committee Member)

Subjects:

Behaviorial Sciences; Environmental Science; Food Science; Oncology; Public Health

Keywords:

black raspberries; oxidative stress; DNA damage; adherence; dimethyl ellagic acid; urolithins; 8-hydroxy-2-deoxyguanosine; food-based; predictive modeling; study participation;

Pinion, Tyson LFactors That Influence Alumni Giving at Three Private Universities
Doctor of Philosophy, University of Toledo, 2016, Higher Education
State and federal funding for higher education is becoming more restrictive at the same time competition for donations to non-profit and educational institutions grows. As such, university development departments are challenged with identifying potential donors and with adopting more efficient practices so as to ensure successful fund-raising campaigns. This study used de-identified alumni donation information from three, private, Ohio-based universities over a 10-year period, 1995-2005. Using Astin’s Theory of Student of Involvement (1984) as its framework, the researcher sought to determine what influence, if any, alumni demographic information, undergraduate fields of study, and undergraduate experiences in on-campus academic, social, and athletic pursuits have on alumni donations. A significant finding from this study is the fact that having alumni involved in more than one on-campus academic, social, or athletic pursuit was the most significant predictor of alumnus total donations, the study’s criterion variable. This study is believed to be the first to have applied Astin’s student involvement theory to alumni donation patterns. Future researchers may identify even more opportunities to target philanthropic opportunities among alumni so as to ensure more efficient, effective higher education donor campaigns.

Committee:

Ronald Opp, PhD (Committee Chair); Debra Harmening, PhD (Committee Member); Michael Coomes, PhD (Committee Member); Jim Troha, PhD (Committee Member)

Subjects:

Higher Education Administration

Keywords:

Alumni; donations; fundraising; philanthropy; student affairs; predictive modeling; alumni affairs; athletics; development; advancement; astin; donor; campaign; target marketing; participation rates; academic affairs; strategic planning; board; private

VENKATESAN, JAYARAMA PATTERN RECOGNITION APPROACH TO POSTURAL STABILITY AND PREDICTION OF WORKPLACE INJURY
MS, University of Cincinnati, 2006, Engineering : Electrical Engineering
Prediction and prevention of falls is a very important part of creating safe and productive workplaces. The risk of falling during workplace activity is related to postural stability. However, assessing postural stability during task performance is difficult and expensive because the task conditions used in testing themselves trigger slips, necessitating the use of strict protocols and protective equipment whereas it would be easier to get data from upright standing static position. Also, using data from such dynamic tests is problematic because the movement of the subjects is likely to make it even more non-stationary than data obtained under static conditions . Thus the research proposed here considers the possibility of using postural data obtained under much simpler static testing conditions to predict postural stability during dynamic workplace tasks.

Committee:

Dr. Ali Minai (Advisor)

Keywords:

postural stability; predictive modeling of postural stability

Schumacher, Ronald MWhat Attracts Students To A Small, Private University?
Doctor of Education (Ed.D.), Bowling Green State University, 2015, Leadership Studies
Few research studies have examined reasons why students choose to attend small, private universities, and even fewer have captured the selection criteria of the students in a qualitative manner (Carr, 2012; Pampaloni, 2010). This study sought to identify reasons why students chose a small, private university, particularly students who chose not to participate in intercollegiate athletes or performing arts. Using a qualitative framework, the researcher conducted 26 semi-structured, in-person student interviews. The findings of this study contribute to the scholarly literature on the topic of enrollment management at private universities in terms of insight and strategies that can increase university enrollment. The researcher sought to explore students’ opinions about factors that were important to them in selecting their institution of choice. Ten primary themes emerged from the interviews: (1) Aspects of affordability and scholarships were important for attending Surreal University (SU). (2) The ability to participate not only in athletics or performing arts programs but also in other activities was significant to attend SU. (3) Academic programs were cited by 20 of the 26 students as very important in their selection process. (4) Students cited the importance of internships that are built into the curriculum at SU. (5) Proximity to the university was important to students. (6) All but 2 of the 26 students interviewed began their college search via the university’s website. (7) Students indicated that the university’s offering of dual enrollment programs at local high schools was important. (8) Students cited local proximity to SU as a reason for their choice primarily because of name recognition and familiarity. (9) Students noted that campus visits contributed to their choosing SU. (10) Finally, students were excited to be able to take part in the study-abroad opportunity or in the university’s honor programs. Given the current outlook for private higher educational institutions in terms of projected decreases in college-age students during the next decade, understanding why students select one institution over another is a valuable tool (a) for determining how and where to invest scarce resources and (b) staying on top of current trends related to the college selection process. Specifically focused on small private institutions that have low to moderately low endowment levels and are very tuition driven in terms of how the institution meets its operational budget goals. This study will enhance previous models for enrollment leaders regarding the importance of creating an enrollment plan based on research of the current student body.

Committee:

Pat Pauken, Dr. (Advisor); Ashutosh Sohoni, Dr. (Other); Mark Early, Dr. (Committee Member); Bonnie Tiell, Dr. (Committee Member)

Subjects:

Higher Education; Higher Education Administration

Keywords:

College Selection; Recruiting; Enrollment Management; Tuition Discounting; Predictive Modeling; Financial Aid

Mintz, Laura JanineAttrition, Translation, and Failure in Interdisciplinary Pain Rehabilitation
Doctor of Philosophy, Case Western Reserve University, 2013, Epidemiology and Biostatistics
This dissertation uses an existing data source from the Cleveland Clinic’s Chronic Pain Rehabilitation Program (CPRP) to describe factors predicting attrition and failure in interdisciplinary pain care, as well as to use the evidence basis from interdisciplinary pain care to translate to concrete policy strategies for use in primary care. The dissertation has three aims: build a model describing factors that patient attrition in an interdisciplinary pain rehabilitation program using the patient data registry at the CPRP, build a nomogram to predict program failure in a comprehensive chronic pain rehabilitation program, using the CPRP patient data registry and examine the evidence basis for interdisciplinary pain care interventions to suggest focused near-term strategies for implementation of the primary care recommendations in the Institute of Medicine’s report Relieving Pain in America. Four factors predictive of program attrition in the CPRP were identified: marital status, IQ, chemical dependence and clinician assessed depression. After imputation, variable selection using Harrrell’s `model approximation’ method and bootstrap validation of the model were used to develop a nomogram to predict program failure. The final nomogram contained ten variables: marital status, IQ, hours rest/day, smoking, chemical dependence, University of Alabama Pain Behavior Scale, anxiety diagnosis, Pain Disability Index, pain duration (years), and the depression subscale of the Depression, Anxiety, and Stress Scale. The model validation results showed the pooled C-statistic=.794 (.722-.803), r2=.226, and shrinkage factor=.97. Interdisciplinary pain care strategies can provide the evidence base for primary care providers to respond to the recommendations from the IOM. Primary care providers and health systems can use evidence based practice to implement these recommendations in the near-term by: changing screening and evaluation of chronic pain, partnering with community stakeholders to provide resources for accessible movement and relaxation, changing physician-patient communication about the disease of chronic pain, educating patients about the neurobiological components of the disease of chronic pain, addressing the behavioral health and environmental components of pain as a standard part of care, connecting chronic pain patients with each other to enhance self-management, and training all providers that work with patients with chronic pain in addiction and opiate management.

Committee:

Kathleen Smyth, PhD (Committee Chair); Duncan Neuhauser, PhD (Committee Member); Kurt Stange, MD/PhD (Committee Member); Michael Kattan, PhD (Committee Member); Judith Scheman, PhD (Committee Member)

Subjects:

Epidemiology; Health; Health Care

Keywords:

health services research; predictive modeling; nomogram; chronic pain; pain management; tertiary care; primary care; interdisciplinary pain

Shewinvanakitkul, PrapanAutomated Detection and Prediction of Sleep Apnea Events
Doctor of Philosophy, Case Western Reserve University, 2017, EECS - System and Control Engineering
This thesis describes a system that can be used for accurate detection and prediction of sleep apnea events by employing signal pre-processing, feature extraction and classification techniques. Sleep apnea syndrome is characterized by the repeated temporary cessation of breathing to the lungs during sleep – an important health problem that can lead to reduced daytime work performance and accidents. Some studies have linked sleep apnea to atrial fibrillation, stroke, myocardial infarction and sudden cardiac death. An important and practical problem is therefore the processing of biomedical signals to extract information that reflect characteristics of sleep apnea. A diagnosis of sleep apnea typically requires full-night multi-channel monitoring by means of overnight polysomnography (PSG). This research addresses the real-time problem of both classifying sleep apnea events and predicting impending apnea events using PSG data available prior to these events. In this study we examined differences between patients who have sleep apnea but are non-hypertensive and those that have sleep apnea but are hypertensive. These patient groups were found to have different characteristics in terms of how they were needed to be handled for accurate detection and prediction of sleep apnea events. Experimental results demonstrate the excellent flexibility of the proposed sleep apnea detection and sleep apnea prediction algorithms in term of accuracy for both groups. As a result, the detection and prediction of individual episodes of sleep apnea is approached using several algorithms that offer promise to reduce health care cost related to treatment.

Committee:

Marc Buchner, PhD (Advisor); Kenneth Loparo, PhD (Committee Member); Vira Chankong, PhD (Committee Member); Frank Jacono, MD (Committee Member)

Subjects:

Artificial Intelligence; Biomedical Engineering; Statistics; Systems Design

Keywords:

Predictive modeling; Classification; Signal processing; Sleep Apnea

Kopicky, Stephen EdwardThe Use of Near Infrared Spectroscopy in Rubber Quantification
Master of Science, The Ohio State University, 2014, Food, Agricultural and Biological Engineering
Taraxacum kok-saghyz and Parthenium argentatum, commonly known as TK and guayule, respectively, are plant species capable of producing high molecular weight natural rubber for use in industrial applications. Cis-1, 4-polyisoprene rubber is a vital natural resource providing physical properties necessary for the manufacturing of over 40,000 consumer products, and not duplicable by synthetic alternatives. Both TK and guayule can be farmed in the U.S, offering alternative sources of natural rubber with prospective economic benefits and improved resource security. Although TK and guayule have potential for domestic cultivation, improvement of agronomic practices and genetic yields are necessary to compete with the current supply of Hevea natural rubber. The objectives of this thesis were to: a) develop near infrared predictive models for domestic rubber crops, allowing rapid quantification of water and rubber content for breeding and experimentation; b) evaluate rubber production and growth of the current TK crop; c) employ hydroponic methods of TK root growth to determine fertilizer regimes for increased growth and rubber production, and to evaluate root clipping harvest methods for hydroponic rubber cultivation. Predictive near infrared models were successfully developed for rubber analysis of ground Taraxacum kok-saghyz roots from multiple sample sources (soil grown, hydroponic, post latex quantification), and water and rubber analysis of Parthenium argentatum through intact bark tissue. Both model types showed high predictive accuracy, with the ground TK model achieving a correlation of 0.89, while the guayule model had correlations of 0.94 and 0.90 for water and rubber, respectively. The use of NIR predictive technology is able to reduce time and resources needed for traditional quantification through accelerated solvent extractions, improving the efficiency of breeding and experimentation. Implementation of a hydroponic system for the growth of TK produced adventitious root masses with high regenerative properties and rubber production comparable to soil grown roots. Hydroponic growth of TK shows promise as a research method, and has potential to be employed as a year round source of root material, through root clipping, for continuous rubber processing. The current Taraxacum kok-saghyz crop contains significant variation in rubber content and plant size, requiring further advancement to develop genetic lines with reliable rubber yields. The methods of growth and rubber quantification developed through this research have the ability to improve scientific techniques for investigation of both species and for agricultural implementation, pushing Taraxacum kok-saghyz and Parthenium argentatum to the forefront of natural rubber cultivation.

Committee:

Katrina Cornish (Advisor); Josh Blakeslee (Committee Member); Peter Ling (Committee Member)

Subjects:

Agricultural Engineering

Keywords:

Natural Rubber; Taraxacum kok-saghyz; Parthenium argentatum; Near Infrared Spectroscopy; Predictive Modeling; Hydroponics; Rubber Cultivation;

Egilmez, GokhanRoad Safety Assessment of U.S. States: A Joint Frontier and Neural Network Modeling Approach
Master of Science (MS), Ohio University, 2013, Civil Engineering (Engineering and Technology)
In this thesis, road safety assessment and prediction modeling for U.S. states fatal crashes are addressed. In the first part, a DEA-based Malmquist Index model was developed to assess the relative efficiency and productivity of U.S. states in decreasing the number of road fatalities. Even though the national trend in fatal crashes has reached to the lowest level since 1949 (Traffic Safety Annual Assessment Highlights, 2010), a state-by-state analysis and comparison has not been studied considering other characteristics of the holistic national road safety assessment problem in any work in the literature or organizational reports. The single output, fatal crashes, and five inputs were aggregated into single road safety score and utilized in the DEA-based Malmquist Index mathematical model. The period of 2002-2008 was considered due to data availability for the inputs and the output considered. According to the results, there is a slight negative productivity (an average of -0.2 percent productivity) observed in the U.S. on minimizing the number of fatal crashes along with an average of 2.1 percent efficiency decline and 1.8 percent technological improvement. The productivity in reducing the fatal crashes can only be attributed to the technological growth since there is a negative efficiency growth is occurred. It can be concluded that even though there is a declining trend observed in the fatality rates, the efficiency of states in utilizing societal and economical resources towards the goal of zero fatality is not still efficient. In the second part, a nonparametric prediction model, Artificial Neural Network, was developed to assist policy makers in minimizing fatal crashes across the United States. Seven input variables from four safety performance input domains while fatal crashes was utilized as the single output variable for the scope of the research. Artificial Neural Networks (ANN) was utilized and the best neural network model was developed out of 1000 networks. The proposed neural network model predicted data with 84 percent coefficient of determination. In addition, developed ANN model was benchmarked with a multiple linear regression model and outperformed in all performance metrics including r, R2 and the standard error of estimate. A sensitivity analysis was also conducted and the results indicated that road length, vehicle miles traveled, and safety expenditures were the top three input variables on fatal crashes. In conclusion, more effective policy making towards increasing safety belt usage and better utilization of safety expenditures to improve road condition are derived as the key areas to focus on for state highway safety agencies from the scope of current research. This research also reveals the significance of the relationship between the four input domains and fatal crashes for the United States from a holistic perspective and offers a robust nonparametric model to policy makers for the prediction of fatal crashes.

Committee:

Deborah McAvoy, Ph.D. (Advisor); Byung-Cheol Kim, Ph.D. (Committee Member); Ken Walsh, Ph.D. (Committee Member); M. Khurrum S. Bhutta, Ph.D. (Committee Member)

Subjects:

Civil Engineering; Industrial Engineering; Transportation

Keywords:

Road Safety Assessment; Benchmarking; Data Envelopment Analysis; Malmquist Productivity Index; Nonparametric Predictive Modeling; Artificial Neural Networks; Machine Learning; US States